An automated tracking system to measure the dynamic properties of vesicles in living cells

SpindlesTracker: An Automatic and Low-Cost Labeled Workflow for Spindle Analysis

Quantitative analysis of spindle dynamics in mitosis through fluorescence microscopy requires tracking spindle elongation in noisy image sequences. Deterministic methods, which use typical microtubule detection and tracking methods, perform poorly in the sophisticated background of spindles. In addition, the expensive data labeling cost also limits the application of machine learning in this field. Here we present a fully automatic and low-cost labeled workflow that efficiently analyzes the dynamic spindle mechanism of time-lapse images, called SpindlesTracker. In this workflow, we design a network named YOLOX-SP which can accurately detect the location and endpoint of each spindle under box-level data supervision. We then optimize the algorithm SORT and MCP for spindle's tracking and skeletonization. As there was no publicly available dataset, we annotated a S.pombe dataset that was entirely acquired from the real world for both training and evaluation. Extensive experiments demonstrate that SpindlesTracker achieves excellent performance in all aspects, while reducing label costs by 60%. Specifically, it achieves 84.1% mAP in spindle detection and over 90% accuracy in endpoint detection. Furthermore, the improved algorithm enhances tracking accuracy by 1.3% and tracking precision by 6.5%. Statistical results also indicate that the mean error of spindle length is within 1 μm. In summary, SpindlesTracker holds significant implications for the study of mitotic dynamic mechanisms and can be readily extended to the analysis of other filamentous objects. The code and the dataset are both released on GitHub.

BNP-Track: A framework for multi-particle superresolved tracking

When tracking fluorescently labeled molecules (termed "emitters") under widefield microscopes, point spread function overlap of neighboring molecules is inevitable in both dilute and especially crowded environments. In such cases, superresolution methods leveraging rare photophysical events to distinguish static targets nearby in space introduce temporal delays that compromise tracking. As we have shown in a companion manuscript, for dynamic targets, information on neighboring fluorescent molecules is encoded as spatial intensity correlations across pixels and temporal correlations in intensity patterns across time frames. We then demonstrated how we used all spatiotemporal correlations encoded in the data to achieve superresolved tracking. That is, we showed the results of full posterior inference over both the number of emitters and their associated tracks simultaneously and self-consistently through Bayesian nonparametrics. In this companion manuscript we focus on testing the robustness of our tracking tool, BNP-Track, across sets of parameter regimes and compare BNP-Track to competing tracking methods in the spirit of a prior Nature Methods tracking competition. We explore additional features of BNP-Track including how a stochastic treatment of background yields greater accuracy in emitter number determination and how BNP-Track corrects for point spread function blur (or "aliasing") introduced by intraframe motion in addition to propagating error originating from myriad sources (such as criss-crossing tracks, out-of-focus particles, pixelation, shot and camera artefact, stochastic background) in posterior inference over emitter numbers and their associated tracks. While head-to-head comparison with other tracking methods is not possible (as competitors cannot simultaneously learn molecule numbers and associated tracks), we can give competing methods some advantages in order to perform approximate head-to-head comparison. We show that even under such optimistic scenarios, BNP-Track is capable of tracking multiple diffraction-limited point emitters conventional tracking methods cannot resolve thereby extending the superresolution paradigm to dynamical targets.

Adaptive tracking algorithm for trajectory analysis of cells and layer-by-layer assessment of motility dynamics

Tracking biological objects such as cells or subcellular components imaged with time-lapse microscopy enables us to understand the molecular principles about the dynamics of cell behaviors. However, automatic object detection, segmentation and extracting trajectories remain as a rate-limiting step due to intrinsic challenges of video processing. This paper presents an adaptive tracking algorithm (Adtari) that automatically finds the optimum search radius and cell linkages to determine trajectories in consecutive frames. A critical assumption in most tracking studies is that displacement remains unchanged throughout the movie and cells in a few frames are usually analyzed to determine its magnitude. Tracking errors and inaccurate association of cells may occur if the user does not correctly evaluate the value or prior knowledge is not present on cell movement. The key novelty of our method is that minimum intercellular distance and maximum displacement of cells between frames are dynamically computed and used to determine the threshold distance. Since the space between cells is highly variable in a given frame, our software recursively alters the magnitude to determine all plausible matches in the trajectory analysis. Our method therefore eliminates a major preprocessing step where a constant distance was used to determine the neighbor cells in tracking methods. Cells having multiple overlaps and splitting events were further evaluated by using the shape attributes including perimeter, area, ellipticity and distance. The features were applied to determine the closest matches by minimizing the difference in their magnitudes. Finally, reporting section of our software were used to generate instant maps by overlaying cell features and trajectories. Adtari was validated by using videos with variable signal-to-noise, contrast ratio and cell density. We compared the adaptive tracking with constant distance and other methods to evaluate performance and its efficiency. Our algorithm yields reduced mismatch ratio, increased ratio of whole cell track, higher frame tracking efficiency and allows layer-by-layer assessment of motility to characterize single-cells. Adaptive tracking provides a reliable, accurate, time efficient and user-friendly open source software that is well suited for analysis of 2D fluorescence microscopy video datasets.

PySTACHIO: Python Single-molecule TrAcking stoiCHiometry Intensity and simulatiOn, a flexible, extensible, beginner-friendly and optimized program for analysis of single-molecule microscopy data

As camera pixel arrays have grown larger and faster, and optical microscopy techniques ever more refined, there has been an explosion in the quantity of data acquired during routine light microscopy. At the single-molecule level, analysis involves multiple steps and can rapidly become computationally expensive, in some cases intractable on office workstations. Complex bespoke software can present high activation barriers to entry for new users. Here, we redevelop our quantitative single-molecule analysis routines into an optimized and extensible Python program, with GUI and command-line implementations to facilitate use on local machines and remote clusters, by beginners and advanced users alike. We demonstrate that its performance is on par with previous MATLAB implementations but runs an order of magnitude faster. We tested it against challenge data and demonstrate its performance is comparable to state-of-the-art analysis platforms. We show the code can extract fluorescence intensity values for single reporter dye molecules and, using these, estimate molecular stoichiometries and cellular copy numbers of fluorescently-labeled biomolecules. It can evaluate 2D diffusion coefficients for the characteristically short single-particle tracking data. To facilitate benchmarking we include data simulation routines to compare different analysis programs. Finally, we show that it works with 2-color data and enables colocalization analysis based on overlap integration, to infer interactions between differently labelled biomolecules. By making this freely available we aim to make complex light microscopy single-molecule analysis more democratized.

Numerical analysis of the impact of cytoskeletal actin filament density alterations onto the diffusive vesicle-mediated cell transport

The interior of a eukaryotic cell is a highly complex composite material which consists of water, structural scaffoldings, organelles, and various biomolecular solutes. All these components serve as obstacles that impede the motion of vesicles. Hence, it is hypothesized that any alteration of the cytoskeletal network may directly impact or even disrupt the vesicle transport. A disruption of the vesicle-mediated cell transport is thought to contribute to several severe diseases and disorders, such as diabetes, Parkinson’s and Alzheimer’s disease, emphasizing the clinical relevance. To address the outlined objective, a multiscale finite element model of the diffusive vesicle transport is proposed on the basis of the concept of homogenization, owed to the complexity of the cytoskeletal network. In order to study the microscopic effects of specific nanoscopic actin filament network alterations onto the vesicle transport, a parametrized three-dimensional geometrical model of the actin filament network was generated on the basis of experimentally observed filament densities and network geometries in an adenocarcinomic human alveolar basal epithelial cell. Numerical analyzes of the obtained effective diffusion properties within two-dimensional sampling domains of the whole cell model revealed that the computed homogenized diffusion coefficients can be predicted statistically accurate by a simple two-parameter power law as soon as the inaccessible area fraction, due to the obstacle geometries and the finite size of the vesicles, is known. This relationship, in turn, leads to a massive reduction in computation time and allows to study the impact of a variety of different cytoskeletal alterations onto the vesicle transport. Hence, the numerical simulations predicted a 35% increase in transport time due to a uniformly distributed four-fold increase of the total filament amount. On the other hand, a hypothetically reduced expression of filament cross-linking proteins led to sparser filament networks and, thus, a speed up of the vesicle transport.

Many vital processes in our eukaryotic cells and organs require an astonishingly precise routing of intermediate products to various intra- and extracellular destinations using vesicles as transporters. This can be illustrated by numerous examples, such as the production and destruction of proteins, the export of neurotransmitters or insulin to the extracellular domain, etc. However, the inside of a cell is tightly packed with numerous structural scaffoldings (filaments), which serve as obstacles and impede the vesicle motion. It is thought that any disturbances of the vesicle-mediated cell transport contribute to numerous degenerative diseases and disorders, which highlights the clinical relevance for investigating this intracellular transport mechanism by developing computational models and performing experimental studies. In this study, we numerically quantified how different specific alterations of the filament density inside a human lung cell—due to changed mechanical loadings or genetic disorders of proteins being responsible for filament branching—affect the diffusion of vesicles inside the intracellular fluid. Therefore, based on the concept of homogenization, a computationally efficient numerical method was developed and utilized to simulate the diffusion of vesicles inside the whole cell, considering the detailed structural information of the filament network.

Gaining insight into cellular cardiac physiology using single particle tracking

Single particle tracking (SPT) is a robust technique to monitor single-molecule behaviors in living cells directly. By this approach, we can uncover the potential biological significance of particle dynamics by statistically characterizing individual molecular behaviors. SPT provides valuable information at the single-molecule level, that could be obscured by simple averaging that is inherent to conventional ensemble measurements. Here, we give a brief introduction to SPT including the commonly used optical implementations, fluorescence labeling strategies, and data analysis methods. We then focus on how SPT has been harnessed to decipher myocardial function. In this context, SPT has provided novel insight into the lateral diffusion of signal receptors and ion channels, the dynamic organization of cardiac nanodomains, subunit composition and stoichiometry of cardiac ion channels, myosin movement along actin filaments, the kinetic features of transcription factors involved in cardiac remodeling, and the intercellular communication by nanotubes. Finally, we speculate on the prospects and challenges of applying SPT to future questions regarding cellular cardiac physiology using SPT.

Assessing particle kinematics via template matching algorithms

Template matching algorithms represent a viable tool to locate particles in optical images. A crucial factor of the performance of these methods is the choice of the similarity measure. Recently, it was shown in [Gao and Helgeson, Opt. Express 22 (2014)] that the correlation coefficient (CC) leads to good results. Here, we introduce the mutual information (MI) as a nonlinear similarity measure and compare the performance of the MI and the CC for different noise scenarios. It turns out that the mutual information leads to superior results in the case of signal dependent noise. We propose a novel approach to estimate the velocity of particles which is applicable in imaging scenarios where the particles appear elongated due to their movement. By designing a bank of anisotropic templates supposed to fit the elongation of the particles we are able to reliably estimate their velocity and direction of motion out of a single image.

A review of progress in single particle tracking: from methods to biophysical insights

Optical microscopy has for centuries been a key tool to study living cells with minimum invasiveness. The advent of single molecule techniques over the past two decades has revolutionized the field of cell biology by providing a more quantitative picture of the complex and highly dynamic organization of living systems. Amongst these techniques, single particle tracking (SPT) has emerged as a powerful approach to study a variety of dynamic processes in life sciences. SPT provides access to single molecule behavior in the natural context of living cells, thereby allowing a complete statistical characterization of the system under study. In this review we describe the foundations of SPT together with novel optical implementations that nowadays allow the investigation of single molecule dynamic events with increasingly high spatiotemporal resolution using molecular densities closer to physiological expression levels. We outline some of the algorithms for the faithful reconstruction of SPT trajectories as well as data analysis, and highlight biological examples where the technique has provided novel insights into the role of diffusion regulating cellular function. The last part of the review concentrates on different theoretical models that describe anomalous transport behavior and ergodicity breaking observed from SPT studies in living cells.

Quantitative comparison of multiframe data association techniques for particle tracking in time-lapse fluorescence microscopy

Biological studies of intracellular dynamic processes commonly require motion analysis of large numbers of particles in live-cell time-lapse fluorescence microscopy imaging data. Many particle tracking methods have been developed in the past years as a first step toward fully automating this task and enabling high-throughput data processing. Two crucial aspects of any particle tracking method are the detection of relevant particles in the image frames and their linking or association from frame to frame to reconstruct the trajectories. The performance of detection techniques as well as specific combinations of detection and linking techniques for particle tracking have been extensively evaluated in recent studies. Comprehensive evaluations of linking techniques per se, on the other hand, are lacking in the literature. Here we present the results of a quantitative comparison of data association techniques for solving the linking problem in biological particle tracking applications. Nine multiframe and two more traditional two-frame techniques are evaluated as a function of the level of missing and spurious detections in various scenarios. The results indicate that linking techniques are generally more negatively affected by missing detections than by spurious detections. If misdetections can be avoided, there appears to be no need to use sophisticated multiframe linking techniques. However, in the practically likely case of imperfect detections, the latter are a safer choice. Our study provides users and developers with novel information to select the right linking technique for their applications, given a detection technique of known quality.

Objective comparison of particle tracking methods

Particle tracking is of key importance for quantitative analysis of intracellular dynamic processes from time-lapse microscopy image data. Because manually detecting and following large numbers of individual particles is not feasible, automated computational methods have been developed for these tasks by many groups. Aiming to perform an objective comparison of methods, we gathered the community and organized an open competition in which participating teams applied their own methods independently to a commonly defined data set including diverse scenarios. Performance was assessed using commonly defined measures. Although no single method performed best across all scenarios, the results revealed clear differences between the various approaches, leading to notable practical conclusions for users and developers.

Morphological filter improve the efficiency of automated tracking of secretory vesicles with various dynamic properties

Membrane trafficking is a very important physiological process involved in protein transport, endocytosis, and exocytosis. The functions of vesicles are strongly correlated with various spatial dynamic properties of vesicles, including their types of movements and morphology. Several methods are used to quantify such dynamic properties, but most of them are specific to particular populations of vesicles. We previously developed the so-called PTrack system for quantifying the dynamics of secretory vesicles near the cell surface, which are small and move slowly. To improve the system performance in quantifying large and fast-moving vesicles, we firstly combined morphological filter with two-threshold image processing techniques to locate granules of various sizes. Next, Kalman filtering was used to improve the performance in tracking fast-moving and large granules. Performance evaluation by using simulation image sequences shown that the new system, called PTrack II, yields better tracking accuracy. The tracking system was validated using time-lapse images of insulin granules in betaTC3 cells, which revealed that PTrack II could track better than PTrack, averaged accuracy up to 56%. The overall tracking results indicate that PTrack II is better at tracking vesicles with various dynamic properties, which will facilitate the acquisition of more-complete information on vesicle dynamics.

Multiple objects tracking in fluorescence microscopy

Many processes in cell biology are connected to the movement of compact entities: intracellular vesicles and even single molecules. The tracking of individual objects is important for understanding cellular dynamics. Here we describe the tracking algorithms which have been developed in the non-biological fields and successfully applied to object detection and tracking in biological applications. The characteristics features of the different algorithms are compared.

Intracellular objects tracking

The tracking of intracellular organelles and vesicles is becoming increasingly important for understanding cellular dynamics. Originally, the development of tracking algorithms was mainly pursued in other fields, e.g. aerospace/military/street surveillance. However, most of this algorithm is not directly applicable to live cell microscopy data. Here we describe the algorithms that have been successfully applied to object detection and tracking specifically in vivo and in vitro motility assays. The characteristics advantages and disadvantages of the different approaches are compared.